Applied Thermal Engineering 141 (2018) 79–89

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Effect of the insulation level on the thermal response of a PCM-modified T

envelope of a dwelling in Chile
⁎
Tomás Venegasa, Diego A. Vascoa, , Fabián E. Garcíab, Carlos Salinasb
a
Departamento de Ingeniería Mecánica, Universidad de Santiago de Chile, Av. Lib. Bernardo O’Higgins 3363, Santiago, Chile
b
Departamento de Ingeniería Mecánica, Universidad del Bío Bío, Av. Collao 1202, Concepción, Chile

H I GH L IG H T S

• Temperature-dependent thermal properties of hexadecane (PCM) were measured.

• PCM-modified walls of a dwelling in two thermal zones of Chile were simulated.
• The effect of the insulation level on the heating and cooling demands was studied.
• IfWhen
the insulation level is increased, the cooling demand is augmented in Santiago.
• the insulation level is increased, the cooling demand is augmented.

A R T I C LE I N FO A B S T R A C T

Keywords: Chile exhibits a continuous growth of energy demand, and for this reason energy saving approaches in the
Thermal envelope commercial, public, and residential sectors, which share 21% of the final energy consumption in Chile, have
Phase change material been encouraged. One of the solutions to increase energy savings in buildings and dwellings is to increase the
Energy demands thermal performance of their envelope. The use of phase change materials (PCMs) has gained attention during
EnergyPlus
recent years, especially in northern climates, since they can be used to enhance the thermal inertia of light
building materials. Usually, the thermal envelope of a dwelling in Chile is made of brick or wood together with
light building materials such like fiber-cement, plasterboard, and thermal insulating materials as polystyrene
foam. The experimental part of this work deals with the thermal characterization of an organic PCM (hex-
adecane), which has a relatively low phase transition temperature. The characterization involves the mea-
surement of density, thermal conductivity, and heat capacity as a function of temperature, and the determination
of phase change temperature and latent heat. The thermophysical properties were implemented in a set of
thermal simulations of an actual project dwelling located in Santiago and Puerto Montt. The numerical results in
terms of heat storage and temperature profiles in the envelope, and heating and cooling demands are shown for
different insulation levels.

1. Introduction
In Chile, the commercial, public and residential sectors share 21%
of the total of energy consumption, with electric energy as the most
important energy source (34%), followed by biomass (32%), LPG
(18%), and NG (11%) [1]. Specifically, the residential sector use 16% of
the total electric energy use in Chile [2]. The average Chilean home has
an annual total energy consumption of 10,232 kWh/year, which is
considered high mainly because of the use of wood as an energy source
in the south of the country, mainly due to its low price and availability
compared to other sources. In southern Chilean cities like Puerto Montt,
wood is used mainly to maintain thermal comfort conditions in houses
[3]. Due to its geographical position with respect to high pressure
zones, the presence of a polar front, and the influence of the Pacific
Ocean and the Andes mountains, Chile has a broad variety of climates.
In 2006 the Ministry of Housing and Urban Development of Chile set
down act 192, which establishes that all housing must meet the require-
ments of thermal conditioning. The details of this act were shown and
discussed in a previous paper [4].
Phase change materials (PCMs) have high latent heat values,
therefore they store large quantities of thermal energy per unit volume.
The heat storage of PCMs as latent heat is three to four times higher
than heat storage as sensible heat [5]. One of the main characteristics of
these materials is that during phase change processes, the temperature

⁎
Corresponding author.
E-mail address: diego.vascoc@usach.cl (D.A. Vasco).

https://doi.org/10.1016/j.applthermaleng.2018.05.083
Received 15 December 2017; Received in revised form 18 May 2018; Accepted 20 May 2018
Available online 26 May 2018
1359-4311/ © 2018 Elsevier Ltd. All rights reserved.
T. Venegas et al.
varies within a narrow range while the material absorbs or releases
thermal energy. PCMs are classified as inorganic, organic and eutectic
mixtures [6]. Among organic PCMs (O-PCMs), three different groups of
substances may be found: paraffins, fatty acids, and organic mixtures
[7]. These materials have melting temperature close to indoor tem-
perature, thus could be potentially used in residential buildings. O-
PCMs are chemically stable, melt and solidify conveniently, without
adding nucleating agents, and therefore they are less prone to sub-
cooling [8]. One of the possible applications of the thermal storage
capacity of O-PCMs is the increase of the thermal inertia of the envelope
materials, the shifting of thermal loads, and the regulation of tem-
perature, helping to decrease temperature variations in a building
during a period of time.
Some authors have studied the inclusion of PCMs in the envelope of
buildings or in additional building components. Omari et al. [9] made
numerical simulations of a one-dimensional heat transfer problem of a
modified insulation material with paraffin PCM particles (100 μ m), and
optimized the thickness of the insulation material and the PCM load.
The effect of melting and solidification of the PCM particles and of the
environmental conditions on the temperature and heat transfer across
the modified material were studied. The authors found that the thermal
conductivity of the insulation material is increased by the addition of
PCM and that the melting temperature is a relevant parameter in the
optimization of the performance of the modified material according to
the season. Moreover, the authors found that the optimal conditions for
summer are not suitable for winter. Hichem et al. [10] performed an
experimental validation of 2D numerical simulations of heat flux
through bricks modified with four different organic PCMs and an in-
organic PCM using ANSYS/FLUENT®. The authors analyzed the effect
of the position of the PCM inclusion on the modified bricks and the type
of PCM on heat flux. Using the inorganic PCM, an 82.1% reduction was
found in the heat flux through the brick. However, only 28% of the
liquid fraction of the PCM was reached. Kuznik et al. [11] developed a
numerical code to simulate a wall containing a commercially available
PCM-enhanced material. The authors analyzed different levels of
thermal insulation in the wall and the thickness of the PCM-enhanced
layer. The results showed that the thickness of the insulation layer does
not affect the optimal PCM layer thickness. Bastani et al. [12] per-
formed a one-dimensional numerical study about the effect of a PCM
wallboard on shifting cooling energy demands. The boundary condi-
tions of the PCM wallboard corresponds to an adiabatic surface and a
time dependent interior set-point temperature. The authors analyzed
ten cases in which the thickness and thermal properties of the PCM
were varied. The study focused on the analysis of the loading cycle of
the PCM wallboard in terms of the thermophysical properties, which
allows the optimization of both PCM and building materials properties
to generate the desired displacement of the peak cooling load and to
reduce the costs of energy consumption.
Other authors have focused on the effects of including PCMs in
building materials on thermal comfort. Kusama et al. [13] prepared and
thermally characterized modified plaster boards with PCM, which were
evaluated in box-shaped test samples and two dwellings with different
heating systems during winter. The authors observed that by including
PCMs, room temperature fluctuation rate is reduced by up to 40%.
Rodriguez-Urbinas et al. [14] studied the effect of the use of micro-
encapsulated PCM (30 wt% and melting temperature of 26 °C) in dry-
wall panels. The authors performed numerical simulations in five cities
with different weather conditions in Spain, analyzing the effect of
window to wall ratio and different shading factors. The results show an
increase of up to 31% of hours within the comfort range by using PCM
materials, compared to reference cases without PCM.
The effect of the inclusion of PCM on the thermal envelope of
buildings on thermal comfort and energy consumption has also been
studied using commercially available building performance simulation
software, which includes modules specifically designed for the simu-
lation of PCM applications. Saffari et al. [15] performed a review of the
80
Applied Thermal Engineering 141 (2018) 79–89
studies that evaluate the possibilities of using PCM to reduce cooling
energy and to increase the number of hours inside the thermal comfort
range in passively cooled buildings. The authors observed that warm
temperature weathers were studied more often due to the potential of
PCM to reduce cooling energy. Soares et al. [16] reviewed studies re-
garding PCM applications in buildings, showing that EnergyPlus,
TRNSYS, and ESP-r are the most widely used software for numerical
simulations of PCM in buildings. EnergyPlus is a commonly used
Building Performance Simulation. By having an incorporated validated
PCM model, EnergyPlus includes the analysis of PCM alongside dif-
ferent energy saving approaches, during the building design stage.
Tabares-Velasco et al. [17] performed a validation of the PCM model
implemented in EnergyPlus following the ASHRAE Standard 140,
which requires an analytic verification, comparative testing, and em-
pirical validation of the numerical results. Panayiotou et al. [18] ana-
lyzed the effect of using PCM enhanced bricks in a dwelling under the
weather conditions of Cyprus on energy consumption and free-floating
temperature during a whole year. The authors performed numerical
simulations using TRNSYS, in which they analyzed different cases with
thermal insulation and PCM element implementation and position. The
results showed that the lower energy consumption was observed in the
case with PCM elements on the exterior surface of the bricks. Moreover,
the case with thermal insulation and PCM elements presented a slightly
higher energy consumption than in the case with only thermal insula-
tion, but a lower free floating temperature during the summer was
observed in the former case. Ramakrishnan [19] performed numerical
simulations of dwellings to evaluate the effect of inner linings of PCM
materials in walls and ceilings for reducing heat stress during heat
waves. The simulations were performed using the CondFD object of
EnergyPlus. The simulations consider a night forced ventilation
strategy. The authors studied a commercially available PCM (Bio-PCM)
and a fictitious PCM with melting temperatures between 25 °C and
31 °C. The authors found that during a heat wave in Melbourne (Aus-
tralia), the use of Bio-PCM along with night ventilation reduced the
hours during which the interior temperature is out of the comfort range
by 32% and up to 62% for the best performing fictitious PCM. Marin
et al. [20] performed numerical studies with EnergyPlus implementing
plaster boards with PCM with a melting temperature of 25 °C in air-
conditioned (packaged terminal heat pumps) containers placed in sev-
eral locations. The authors found that the energy savings in air con-
ditioning depends on the weather conditions. Of the assessed locations,
the highest reductions in cooling and heating loads were observed in
Calama (Chile). Furthermore, the authors evaluated the free floating
temperatures in the containers at the same locations, and found that the
use of PCM increases the number of hours within the thermal comfort
range, except in tropical locations. Chernousov et al. [21] studied nu-
merically with EnergyPlus the effect of implementing PCM elements in
an office building envelope located in a subtropical climate to reduce
cooling energy during the summer morning hours. The study con-
sidered two stages. The first one consisted in simulations to determine
the effects of PCM position within the envelope, the amount of solar
radiation received and PCM layer thickness. After these analyses, the
authors studied the PCM enhanced PCM building including the chiller
of the cooling plant, to evaluate the effect of the PCM on cooling energy
consumption. The results showed that using the PCM on the interior
surfaces of the building reduces the energy consumption by 1% to 4%,
and that increasing the thickness of the PCM layer increases cooling
energy consumption.
The present work is focused first on the thermal characterization of
hexadecane as a phase change material of a relative low phase transi-
tion point (18 ° C), near to the transition point of some fatty acids [22],
which may be of interest in near to room temperature applications
mainly thanks to their condition of no oil-derived product. Thermal
conductivity, heat capacity, and density were measured as a function of
temperature, in both the liquid and the solid phase. These properties
are required in the second part of the paper that deals with the thermal
T. Venegas et al. Applied Thermal Engineering 141 (2018) 79–89

simulation with EnergyPlus of an actual dwelling project in Chile to be 805

built. Different levels of thermal insulation along with a layer of the 798
PCM on the exterior walls were analyzed. The thermal simulations were
791

ȡ [kJ/m3]
performed in two different thermal zones of Chile, Santiago (Zone 3),
and Puerto Montt (Zone 6). Both cities are important urban centers in 784
Chile, and they were selected because of their significant latitude dif- 777
ference and therefore climate conditions. The numerical results in terms
770
of cooling and heating power, stored heat in the envelope, and heating
and cooling demands of the dwelling are analyzed. 763
282 285 288 291 294 297 300 303 306
T [K]
2. Thermophysical properties of hexadecane
Fig. 2. Density of pure hexadecane.

The experiments were performed with reagent grade hexadecane of

7000
99% purity (Sigma–Aldrich). Both specific heat and latent heat were
measured by the DSC technique (TA 2920 Modulated) in the tem- 6000
perature range of 283–313 K. The thermal conductivity was determined
5000
by the transient line source method (KD2-Pro), which consists in the

Cp [J/kgK]
immersion of a sensor (SH-1 needle) that generates heat from its sur- 4000
face. The fluid absorbs a fraction of this heat and the remainder in- 3000
creases the surface temperature. The temperature of the sensor in
2000
comparison to the added heat allowed the measurement of the thermal
conductivity of the fluid of interest. The measurement of the density of 1000
hexadecane was determined by using a pycnometer (Marinfield) with a 0
volume of 9.869 mL. The pycnometer was equipped with a mercury 282 286 290 294 298 302 306 310
thermometer (± 0.1 K), which allowed the measurement of the tem- T [K]
perature of the sample. The temperature was controlled by immersion Fig. 3. Heat capacity of pure hexadecane.
of the hexadecane sample in a thermoregulated water bath (Hilab BL-
20) for the measurement of thermal conductivity (283–313 K) and
density (283–307 K). The measurements of thermal conductivity and latent heat of 216 kJ/kg is obtained from the DSC curve tracing a base
density were performed in triplicate. line and calculating the area under the curve. This value is 8% lower
The results in the temperature range of 10–40 °C for the thermal than that reported by Vélez et al. [23]. Then specific enthalpy as a
conductivity of hexadecane are shown in Fig. 1. In the solid zone (I), the function of temperature was obtained from the heat capacity data using
thermal conductivity may be considered nearly constant at 0.285 W/ Eq. (1), and the obtained results of enthalpy are shown in Fig. 4.
mK, and in the liquid zone (III) the thermal conductivity is lower and T ⩽ Tm T2
almost constant at 0.143 W/mK. These values are in agreement with h= ∫T 1
Cps (T ) dT + Lpc + ∫T ⩽Tm
Cpl (T ) dT
(1)
those reported by Vélez et al. [23].
Fig. 2 shows the density variation as a function of temperature. In Despite of their relative low melting temperature, in comparison for
the solid (I) and liquid (II) zones, density decreases almost linearly with instance with octadecane, hexadecane is more chemically stable and
temperature according to the equation ρ = α + βT , where α is equal to therefore less prone to ageing. Another advantage is its lower cost next
−0.89 and −0.67 for the solid and liquid zone, respectively. Mean- to octadecane. When performing transient thermal simulation the
while, β is equal to 1051.6 and 968.6 for the solid and liquid zone, thermal properties as a function of temperature are required. The per-
respectively. The implemented techniques for the determination of formed simulations are based on the thermal balance method that al-
thermal conductivity and density did not allow a more precise de- lows the instantaneous sensible load to be calculated for the air in a
termination of these properties in the narrow mushy zone. zone.
The values obtained for the heat capacity as a function of tem-
perature are shown in Fig. 3, it can be seen the heat capacity of hex-
adecane presents one peak at 17 °C (II). In the solid region (I), the heat
capacity of hexadecane increases with temperature, and such increase
turns steeper as the temperature gets closer to the mushy zone (II).
Meanwhile, in the liquid region (III), the heat capacity decreases with
temperature to 25.5 ° C and then it increases to 29 °C, and from this
temperature the property becomes almost constant (IV). Finally, a

0.35

0.3
k [W/mK]

0.25

0.2

0.15

0.1
282 286 290 294 298 302 306 310 314
T [K]
Fig. 1. Thermal conductivity of pure hexadecane. Fig. 4. Enthalpy of pure hexadecane.

81
T. Venegas et al. Applied Thermal Engineering 141 (2018) 79–89

3. Simulation tool and numerical formulation

The computational analysis proposed in this paper is based on

building computational simulations using EnergyPlus, which is a well-
recognized and accepted building energy analysis software tool
[24,25]. EnergyPlus has been used by engineers, architects and re-
searchers to model energy consumption (heating, cooling, ventilation)
in buildings. To obtain the transient temperature values within a zone,
EnergyPlus implements an energy balance equation for each zone (z).
The transient term of the equation is the thermal energy stored in zone
air, and the other terms of the balance corresponds to: (i) sum of the
convective internal loads, (ii) convective heat transfer from the zone
surfaces, (iii) heat transfer due to infiltration of outside air, (iv) heat
transfer due to inter-zone air mixing, and (v) air systems output.
The main goal of the present work is to evaluate the thermal per-
formance of a PCM-enhanced projected dwelling under two different
weather conditions in Chile and different thermal insulation levels. No
actual HVAC equipment was considered, and instead an ideal load
system has been included to determine the instantaneous heating and
cooling loads. For the calculation of the transient variation of tem-
perature in the thermal envelope of a building, EnergyPlus implements
a finite difference discretized version of the conduction transfer equa-
tion. For the formulation of the transient term, the Crank–Nicholson
scheme, which it is second-order in time, was implemented. To simulate Fig. 5. Location of both Chilean cities.
PCM on EnergyPlus, objects detailing the enthalpy and thermal con-
ductivity as a function of temperature have to be included. These ob-
to default values provided by the DesignBuilder software, and the
jects are called MaterialPropertyPhaseChange and
lighting power density value corresponds to the living quarters value of
MaterialPropertyVariableThermalConductivity, respectively. In the
ASHRAE standard 90.1–2007 (Table 9.6.1) [27].
MaterialPropertyPhaseChange, a set of data points of enthalpy of the
The SurfaceConvectionAlgorithm:Inside object was used to obtain
PCM as a function of temperature have to be incorporated, and
the indoor convection coefficient. This model defines the convection
EnergyPlus interpolates between the given points. The same procedure
heat transfer coefficient for all inside surfaces. In this case, the TARP
is used by the MaterialPropertyVariableThermalConductivity object to
algorithm has been selected, which is the default model in EnergyPlus
describe thermal conductivity. The values of the temperature-depen-
and combines natural and wind-driven convection correlations from
dent enthalpy and thermal conductivity used in the simulation were
laboratory measurements on flat plates. The convective coefficients
obtained from the experimental results shown in Section 2.
used for the calculation of the thermal transmittance for the outer en-
vironment were obtained using the DOE-2 model, based on the ex-
4. Description of the dwelling
perimental measurements performed by Kleims and Yazdanian for
roughened surfaces [28]. DOE-2 is based on the MoWiTT and BLAST
Due to its geographical position, the presence of a polar front and
models [29,30]. The details of the calculation of the outer convective
the influence of the Pacific Ocean and the Andes mountains, Chile has a
coefficient was described in a previous work [4].
wide variety of climates. The elevation of the western Coast mountain
The outdoor radiative transfer is calculated as two different terms,
chain does not allow the flow of the sea winds and the Andes mountains
short wavelength, which models direct and diffuse radiation heat flux,
inhibit the influence of the continental climate, therefore cities like
and long wavelength that models radiation flux exchange with the sky
Santiago (Csb) has a dry Mediterranean climate. The long coast and the
and surroundings. Short wavelength is influenced by location, the
Humboldt current turn the climate of the country in predominantly
surface facing angle and tilt, surface face material properties, and
marine, which moderates temperature through the development of
weather conditions. Short wavelength comprises beam radiation from
clouds and fresh winds. Puerto Montt (Cfb) has a rainy marine climate
the sun and diffuse radiation from the sky. Radiation from the sky in-
with lower average temperatures and higher rainfall than in Santiago.
cludes beam radiation, diffuse anisotropic from the Skydome, and
The TMY3 weather data were obtained from the EnergyPlus webpage
concentrated horizon radiation. The external longwave radiation be-
for Santiago [26]. For Puerto Montt, it was used a Chilean weather
tween surface, sky and the ground is influenced by surface absorptivity,
database generated by the software Meteonorm. The location of both
surface temperature, sky and ground temperatures and sky and ground
cities is shown in Fig. 5.
view factors. The inside radiation is comprised of two different ele-
The project dwelling under simulation is made up of ten zones, and
ments: shortwave radiation, which is the radiation that reaches the
it has a built area of 183 m2 (Fig. 6). These kind of dwellings are be-
internal surfaces of the building through the windows, and longwave
coming very common in the Chilean housing project market, even in
radiation that accounts for absorption and emittance of low-tempera-
cities of different climates like Puerto Montt and Santiago. The dwelling
ture radiation surfaces such as other surfaces or equipment. EnergyPlus
has one floor below ground and two floors above ground, with 8% of
uses a grey interchange model for the longwave radiation among zone
the walls below ground. The window to wall ratio is 19%, windows are
surfaces. This model is based on the concept of ScriptF that relies on a
double glazed with a thermal transmittance of 1.7 W/m2K and a solar
matrix of exchange coefficients between pairs of surfaces that include
heat gain coefficient (SHGC) of 0.59. The structural elements of the
all exchange paths between the surfaces, considering that all radiation
dwelling (walls and slabs) are made of concrete. Thermal insulation is
surfaces are grey and all radiation is diffuse. For more details about the
implemented in all the dwelling surfaces. The configuration and the
modeling of radiation refers to the documentation of EnergyPlus [29].
thermal transmittance values of the walls, floors and roof, according to
the Chilean thermal regulation, are shown in Tables 1 and 2, respec-
tively. Each zone of the dwelling has internal loads, which are specified
in Table 3. The appliance power density and the occupancy correspond

82
T. Venegas et al. Applied Thermal Engineering 141 (2018) 79–89

Fig. 6. Blueprint of the dwelling under simulation with EnergyPlus with an orientation rotated 110° respect to the North.

Table 1 Table 2
Configuration of the thermal envelopes of the dwelling. Thermal transmittance [W/m2K] of the thermal envelope of the dwelling in the
different analyzed cases (TR: Chilean Thermal Regulation).
Thermal Material Thickness k [W/mK] ρ [kg/ Cp [J/
envelope [cm] m3] kgK] Thermal Envelope Project TR Santiago TR Puerto Montt

Below grade Brickwork 10 0.890 1920 790 Roof 0.303 0.47 0.28
wall XPS 7.95 0.04 15 1400 Walls 0.501 1.9 1.1
Concrete 10 1.4 2100 840 Below ground walls 0.356 0.356 0.356
Gypsum 1.3 0.25 900 1000 Floor 0.571 0.571 0.571
plaster Windows 1.786 1.786 1.786

Wall Cement 1
plaster
XPS 5 Table 3
Concrete 20 Internal loads of the dwelling.
Ground floor EPS 5 0.034 35 1400 Zone Appliances power Lighting power Occupancy [W/
Concrete 10 1.95 2240 900 density [W/m2] density [W/m2] m2 ]
Tiles 1.5 0.890 1920 790
Bedroom 5.00 12.00 7.14
External floor Plaster 0.025 0.25 900 1000
Hall 1.57 12.00 64.5
Stone wool 14.8 0.04 30 840
Basement 3.58 12.00 43.5
Timber 0.5 0.119 1636 447
Kitchen 30.28 12.00 42.19
Roof Clay tile 1 Sitting room 12.00 12.00 0
Concrete 5 Living room 12.00 12.00 0
EPS 12
Concrete 20

83
T. Venegas et al. Applied Thermal Engineering 141 (2018) 79–89

Table 5
Area of the windows (SHGC=0.598) of the dwelling and their orientation.
Window Area [m2] Azimuth [° ] Cardinal orientation

1 2.58 20 N
2 4.83 20 N
3 5.64 200 S
4 0.28 290 W
5 3.32 290.15 W
6 3.62 200 S
7 5.97 20 N
8 6.73 20 N
9 2.25 200 S
10 1.76 110 E
Fig. 7. Wall configuration of the outer walls of the dwelling. 11 0.28 290 W
12 0.28 290.12 W
13 2.95 110 E
5. Results and discussion 14 0.56 290 W
15 1.68 200 S
The effect of the thickness of the thermal insulation on the perfor- 16 1.68 110 E
17 0.2 110 E
mance of a layer of pure hexadecane (10 mm) inserted in the walls 18 0.28 110 E
(Fig. 7) was analyzed from a minimum thermal insulation thickness of 19 0.61 20 N
0.85 mm and 12 mm (case 2), according to the Chilean Thermal Reg- 20 5.88 20 N
ulation (TR) for Santiago and Puerto Montt, respectively, to a maximum
value of 50 mm implemented in the project dwelling (case 7). Each of
the additional cases with PCM (cases 2 to 7) corresponds to the dwelling Table 6
with five different levels of thermal insulation (Table 4). Finally, cases 1 Stored thermal energy in some elements of the thermal envelope of the project
and 8 correspond to the baseline case according to the thermal reg- dwelling.
ulation and the project house, both without PCM. Thermal Envelope kWh/ m2 of built area %
Solar energy gains through windows are a relevant element of the
Walls above ground 0.973 70.02
annual energy balance of most buildings. They are especially important
Walls below ground 0.0450 3.24
in housing because houses tend not to have high internal loads, Roof 0.290 20.90
therefore besides mechanical heating, solar gains through windows Ground floor 0.0728 5.24
could be the main energy gain of houses. By this reason, it is important Exterior floor 0.00852 0.61
to appropriately design the windows and its external sun control ele- PCM 0.728 52.42

ments to provide adequate solar gains during the colder season and to
reduce the solar gains during the warmer period. The area of the win-
dows of the project dwelling and their orientation are shown in Table 5.
The project dwelling has already a significant amount of thermal
mass in its envelope. The sensible heat storage capacity of the main
parts of the envelope (per built area of the dwelling), considering a
temperature gradient of 5 °C, have been calculated and compared to the
latent heat store capacity of the PCM (Table 6). It may be seen how the
energy stored as sensible heat is 20% higher than the energy stored as
latent heat in the PCM, mainly due to the contribution of the concrete
layer (0.809 kWh/ m2 of built area), but it is important to clarify that the
mass of the concrete layer is almost six times higher than the mass of
PCM.
Finally, to the better understanding of the results presented in this
section, it is performed an energy analysis of the project dwelling (case
8) in its original location (Santiago), to evaluate the importance of the
different energy gains (Fig. 8). The main sources of energy gains of the
dwelling are the internal ones (equipment, lighting, occupancy) and the
solar gains (R) through the windows, which is indeed the most
Fig. 8. Annual energy gains of the project dwelling located in Santiago. C:
Table 4 Convection, and R: solar gains.
Studied cases of the dwelling with different thickness of thermal insulation
located in Santiago (SCL) and Puerto Montt (PMC).
important, sharing a 58% of the total energy gains.
Case 1 2 3 4 5 6 7 8

U [W/m2K], 1.908 1.908 0.833 1.103 1.37 1.633 0.501 0.501

5.1. Cooling and Heating power
SCL
U [W/m2K], 1.14 1.14 0.676 0.785 0.887 0.991 0.501 0.501
PMC The cooling (Zone Ideal Loads Zone Total Cooling Rate) and heating
Insulation 0.85 0.85 23.0 13.0 7.0 3.0 50 50 power (Zone Ideal Loads Zone Total Heating Rate) are obtained directly
[mm], SCL by EnergyPlus when the user explicitly asks for these variables to be
Insulation 12.0 12.0 32.5 25.5 20.5 16.5 50 50
written in the output file. The cooling power is the total (sensible plus
[mm], PMC
PCM layer No Yes Yes Yes Yes Yes Yes No latent) cooling power delivered to the zone, and the heating power is
the total (sensible plus latent) heating power delivered to the zone. The

84
T. Venegas et al.
Fig. 9. Cooling power demands during the week of the warmer day (December
20) in Santiago. The detail shows the higher cooling power required in cases 7
and 8.
following results show the thermal power demand for cooling and
heating of an ideal air conditioning system. The cooling and heating
power demands should not be confused with the cooling and heating
power consumption of an actual air conditioning system.
Figs. 9 and 10 represent the variation of the cooling power demand
during the week with the warmer day for each location. For Santiago
and Puerto Montt, this day corresponds to December 20 and February
22, respectively. Broadly speaking, the cooling power demand values
exhibit the same behavior for both locations, showing minimum cooling
demand values during the early hours of the day and during the mid-
day hours. On the contrary, the higher values of cooling demands are
observed during the late afternoon and the evening. It is seen that the
cooling demands are higher in Santiago than in Puerto Montt, espe-
cially from 16:00 h to 22:00 h. Moreover, the effect of the different
Fig. 10. Cooling power demands during the week of the warmer day (February
22) in Puerto Montt. The detail shows the higher cooling power required in
cases 7 and 8.
85
Applied Thermal Engineering 141 (2018) 79–89
Fig. 11. Heating power demands during the week of the colder day (June 2) in
Santiago. The detail shows the higher heating power required in cases 2 and 3.
thermal insulation levels on the cooling power demand in Santiago is
more noticeable. In this analysis, case 1, corresponding to the case with
thermal insulation according to the Chilean thermal regulation, is
considered the baseline case. It is seen than the cooling power demands
in the dwelling located in Santiago are higher in case 2 than in case 1,
which is explained by having the minimum thermal insulation level and
a PCM that stores the maximum amount of energy during the sunlight
hours. The results show that case 2, corresponding to PCM-modified
walls and the same level of thermal insulation as the baseline case 1,
presents an increase of maximum cooling power demand of 15.5% and
2.4% in the cities of Santiago and Puerto Montt, respectively. Com-
paring the case for the maximum level of thermal insulation (Case 7:
Project) in Santiago de Chile with both baseline case (1) and case 2, it is
seen that the maximum cooling power demand is reduced by 10.5% and
22.5% with respect to case 1 and case 2, respectively. The same com-
parison for the city of Puerto Montt shows an increase of the maximum
cooling power demand of 9.9% and 7.3% with respect to case 1 and
case 2, respectively.
The heating power demands for both locations are shown in Figs. 11
and 12. It is seen that the minimum heating power demands take place
during the early hours of the day and during mid-day, and the max-
imum heating power demands take place during the morning and after
18:00 h to 23:00 h. The maximum heating demands are seen in Santiago
and the effect of the thermal insulation level is more noticeable in both
locations than in the former analysis. In the results corresponding to
Santiago, it is seen that the heating power demands are higher for the
dwelling with the minimum level of insulation and PCM-modified walls
(case 2), and this effect is reduced by increasing the level of thermal
insulation up to that of the project level (case 7). This trend may be
explained by the reduction of heat loss to the exterior by the increased
thermal insulation. In Puerto Montt a different behavior is observed.
The maximum heating power demands correspond to the baseline case,
though the heating power demand decreases as the thermal insulation
level is increased. In comparison with the baseline case, it was found
that case 2 presents a maximum increase and reduction of the max-
imum heating power demand of 32% and 18.7% in the cities of San-
tiago and Puerto Montt, respectively. Comparing the project case 7 in
Santiago with both baseline cases 1 and 2, it seen that the maximum
heating power demand is reduced by 58.4% and 68.5% with respect to
case 1 and case 2, respectively. The same comparison for the city of
Puerto Montt shows an increase of the maximum cooling power
T. Venegas et al.
Fig. 12. Heating power demands during the week of the colder day (June 5) in
Puerto Montt. The detail shows the higher heating power required in cases 2
and 3.
demand of 55.9% and 45.8% compared to cases 1 and 2, respectively.
The outdoor variation of the temperature in both location is shown in
Figs. 9–12. The inner temperature is maintained almost constant at
25 °C and 15 °C during summer and winter, respectively.
The heat storage (Surface Heat Storage Energy) is the difference
between the inside and outside face conduction, with the sign con-
vention arranged so that positive values indicate heat being added to
the core of the surface. In Figs. 13 and 14, it is seen that the heat stored
in the walls of the dwelling during the summer day reaches its peak
during the morning, after a fast increase coincident with the sunlit
hours. In Santiago, after the peak of heat storage, the amount of stored
energy is reduced hourly, and depending on the level of thermal in-
sulation, it ceases between 16:00 and 18:00 h. After this period of
Fig. 13. Stored energy in the walls during the week of the warmer day
(December 20) in Santiago.
86
Applied Thermal Engineering 141 (2018) 79–89
Fig. 14. Stored energy in the walls during the week of the warmer day
(February 22) in Puerto Montt.
accumulation, the walls start to discharge, losing part of the stored
heat. The peak of discharge occurs at 20:00 h for the different levels of
insulation. After this point, the hourly rate of energy loss by the walls is
reduced at a slow rate. It can be seen that the case with the higher level
of insulation and PCM (case 7) has the lower values in both accumu-
lation and discharge. The case with the lowest insulation level and PCM
(case 2) presents the highest levels of both accumulation and discharge
rates, even higher than the values of the case with the same insulation
level but without PCM (case 1). This observation maybe be explained
by the fact that the inclusion of PCM increases the heat capacity and
therefore the amount of heat that can be stored in the walls, which is
reduced by increasing the amount of thermal insulation, as expected. A
similar behavior was observed in Puerto Montt, but in this case the
absolute amount of energy stored in the walls is lower than in Santiago,
which is attributable to the lower temperatures of Puerto Montt. It can
be seen that the differences between the cases are lower than those
founded in Santiago. Another difference between Puerto Montt and
Santiago is that after the discharge peak (approximately at 20:00 h), the
level of discharge is reduced for all cases after 21:00 h. The same be-
havior between cases 1 and 2 may be observed in Puerto Montt, where
the case with the lower insulation increases its level of both accumu-
lation and discharge by including PCM in the walls (case 2), but this
effect is reduced by increasing the level of insulation, and by reaching
the highest level of insulation analyzed (case 7), the lower peaks of
accumulation and discharge can be seen.
During winter in both locations (Figs. 15 and 16), the highest
amount of stored energy during the colder day is lower than during the
warmer summer days. Another common observation for both locations
is that the time of highest level of accumulation is shifted from the
morning during summer (10:00 h) to the afternoon (14:00 h in Santiago
and 13:00 h in Puerto Montt). In Santiago, the walls stop losing heat
and start accumulating energy at approximately 9:00 h, increasing their
level of energy accumulation in a slower way than during summer,
reaching the peak at 14:00 h, after which the walls start accumulating
less energy hourly until 17:00 h, when they start releasing the energy
accumulated during the day. The same as during the summer, the case
with the lowest variations in accumulation and discharge is that with
PCM and the highest level of insulation (case 7), and the case with the
highest amplitude between accumulation and discharge is the case with
T. Venegas et al.
Fig. 15. Stored energy in the walls during the week of the colder day (June 2)
in Santiago.
Fig. 16. Stored energy in the walls during the week of the colder day (June 5)
in Puerto Montt.
the lowest level of insulation and PCM (case 2). For Puerto Montt, the
same period of accumulation as in Santiago may be observed (between
9:00 h and 17:00 h), but the same as in summer, the different cases
present a more similar behavior between them than in Santiago. In
Puerto Montt, it can also be seen that the peak energy accumulation has
absolute values similar to those of the discharge, which does not hap-
pens in the other analyzed cases, even though the peak during dis-
charge is reduced faster. Except for the peak in discharge at 18:00 h, the
case with the highest insulation and PCM (case 7) has the most stable
behavior between accumulation and discharge, in contrast with the
case with the lowest level of insulation and PCM (case 2), which pre-
sents a greater accumulation and discharge than the other cases.
87
Applied Thermal Engineering 141 (2018) 79–89
Fig. 17. Annual heating and cooling demands for both location: Santiago (up)
and Puerto Montt (down).
5.2. Energy demands of the dwelling
The Energy demand is obtained by calling for two variables, cooling
energy (Zone Ideal Loads Zone Total Cooling Energy) and heating en-
ergy (Zone Ideal Loads Zone Total Heating Energy) to the output file of
EnergyPlus. The annual heating and cooling demands for both locations
are shown in Fig. 17. In general, it is seen that the heating and cooling
demands decrease and increase, respectively, when the level of thermal
insulation is augmented, respectively. Moreover, when a layer of PCM
is added to the envelope according to the thermal regulation (case 2)
the heating and cooling demands are increased and decreased com-
pared to the baseline case 1. With respect to the cooling demands in
Santiago and Puerto Montt, case 2, i.e., thermal envelope according to
the thermal regulation and a PCM layer, show a reduction of the energy
demand of 8.3% and 28.7% compared to the baseline case 1 (thermal
envelope according to the thermal regulation without a PCM layer),
respectively. Meanwhile, the project case 7 presents an increase of
47.8% (Santiago) and 107.3% (Puerto Montt) of the cooling demand,
compared to the same baseline case 1. When a layer of PCM is in-
corporated in the thermal envelope in compliance with the thermal
regulation (case 2), the heating demands are increased by 64.9% and
43.7% compared to the baseline case 1 for the locations of Santiago and
Puerto Montt, respectively. But if the level of insulation is increased to
the project value (case 7) the heating demand is reduced by 83.5% and
T. Venegas et al.
59% in comparison to the baseline case for Santiago and Puerto Montt,
respectively. Regarding the total demands (heating and cooling), the
study in Santiago shows that case 7 and case 2 present lower (5.7%) and
higher (21.4%) values compared to the baseline case 1, respectively.
The same analysis in Puerto Montt shows a decrease of 9% (case 7) and
an increase of 21.9% (case 2) of the total demands compared to case 1.
The obtained results of the energy demand are of the same order of the
average annual total energy consumption of 10,232 kWh/year, but
these results do not consider the electrical consumption of household
appliances. In comparison to the average house in Chile, the energy
demands for air conditioning of the project dwelling are higher given its
relative large built area.
6. Conclusions
The effect of the insulation level of a PCM-modified thermal en-
velope of a project dwelling located in Santiago and Puerto Montt on
the cooling and heating power demands, stored energy in the envelope,
and energy demands was analyzed. With respect to the cooling power
demands, the effect of the insulation level is only noticeable in the case
of the dwelling located in Santiago during the night hours, showing that
the cooling demands decrease with the insulation level. It is important
to remark that the weather in Santiago during summer is warmer than
in Puerto Montt. When the heating demands are analyzed, a more no-
ticeable effect of the insulation level is seen. In Santiago, it may be
noticed that the heating power demand decreases as the insulation level
is increased, while in Puerto Montt the same trend is observed but to a
lower extent.
The energy stored in the walls during the warmer and colder days
was also analyzed in both locations. The effect of the insulation level is
more noticeable in the dwelling located in Santiago, especially during
the warmer summer day. The incorporation of PCM to the thermal
envelope has an increasing effect in the amount of both stored and
released heat during the warmer and colder hours of the day, respec-
tively. The effect of the insulation level in Puerto Montt is less notice-
able since the level of insulation in this city is already high, according to
the Chilean Thermal Regulation.
The addition of PCM to the thermal envelope according to the
Chilean Thermal Regulation, increases and decreases the heating and
cooling energy demand, respectively. Specifically, the cooling energy
demand is decreased in both Santiago (8%) and Puerto Montt (29%),
while the heating demand is increased by 64% and 43% in Santiago and
Puerto Montt, respectively. As the level of insulation is increased, the
heating demand is reduced and the cooling demand is increased, which
is more evident in the dwelling located in Santiago.
According to these results, a PCM with a relatively low phase
change temperature, like hexadecane works, better for cooling pur-
poses, especially in Puerto Montt. In a future work an interesting study
may be the incorporation of two different PCMs, with low and high
phase change temperatures, in the thermal envelope, analyzing the
effect over the cooling and heating demands of each PCM and different
proportions of both in a combined element.
Acknowledgements
This work was funded by CONICYT – Chile under Fondecyt Project
11130168.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.applthermaleng.2018.
05.083.
88
Applied Thermal Engineering 141 (2018) 79–89
References
[1] M. de Energía, Energía 2015: Política energética de chile, Tech. Rep., Ministerio de
Energía: Gobierno de Chile, 2016.
[2] C.A. Agostini, M.C. Plottier, E.H. Saavedra, La demanda residencial de energía
eléctrica en chile, Economía chilena 15 (3) (2012) 64–83.
[3] Estudio de usos finales y curva de oferta de la conservación de la energía en el sector
residencial, Cámara chilena de la construcción, 2010. URL http://antiguo.
minenergia.cl/minwww/export/sites/default/05_Public_Estudios/descargas/
estudiosUsos_Finales_COC_Sector_Residencial_2010.pdf.
[4] D.A. Vasco, M. Munoz-Mejías, R. Pino-Sepúlveda, R. Ortega-Aguilera, C. García-
Herrera, Thermal simulation of a social dwelling in chile: Effect of the thermal zone
and the temperature-dependant thermophysical properties of light envelope mate-
rials, Appl. Therm. Eng. 112 (2017) 771–783, http://dx.doi.org/10.1016/j.
applthermaleng.2016.10.130 URL http://www.sciencedirect.com/science/article/
pii/S1359431116326321 .
[5] H. Mehling, L. Cabeza, Heat and Cold Storage with PCM: An Up to Date
Introduction Into Basics and Applications, Springer, Berlin Heidelberg, 2008.
[6] B. Zalba, J.M. Marn, L.F. Cabeza, H. Mehling, Review on thermal energy storage
with phase change: materials, heat transfer analysis and applications, Appl. Therm.
Eng. 23 (3) (2003) 251–283, http://dx.doi.org/10.1016/S1359-4311(02)00192-8
URL http://www.sciencedirect.com/science/article/pii/S1359431102001928 .
[7] A. Abhat, Low temperature latent heat thermal energy storage: Heat storage ma-
terials, Sol. Energy 30 (4) (1983) 313–332, http://dx.doi.org/10.1016/0038-092X
(83)90186-X URL http://www.sciencedirect.com/science/article/pii/
0038092X8390186X .
[8] V.V. Tyagi, D. Buddhi, Pcm thermal storage in buildings: A state of art, Renew.
Sustain. Energy Rev. 11 (6) (2007) 1146–1166, http://dx.doi.org/10.1016/j.rser.
2005.10.002 URL http://www.sciencedirect.com/science/article/pii/
S1364032105000973 .
[9] K.E. Omari, Y.L. Guer, P. Bruel, Analysis of micro-dispersed PCM-composite boards
behavior in a building’s wall for different seasons, J. Build. Eng. 7 (2016) 361–371,
http://dx.doi.org/10.1016/j.jobe.2016.07.013 URL http://www.sciencedirect.
com/science/article/pii/S2352710216301115 .
[10] N. Hichem, S. Noureddine, S. Nadia, D. Djamila, Experimental and numerical study
of a usual brick filled with PCM to improve the thermal inertia of buildings, Energy
Procedia 36 (2013) 766–775, http://dx.doi.org/10.1016/j.egypro.2013.07.089
URL terraGreen 13 International Conference 2013 – Advancements in Renewable
Energy and Clean Environment http://www.sciencedirect.com/science/article/
pii/S1876610213011752 .
[11] F. Kuznik, J. Virgone, J. Noel, Optimization of a phase change material wallboard
for building use, Appl. Therm. Eng. 28 (11) (2008) 1291–1298, http://dx.doi.org/
10.1016/j.applthermaleng.2007.10.012 URL http://www.sciencedirect.com/
science/article/pii/S135943110700350X .
[12] A. Bastani, F. Haghighat, J. Kozinski, Designing building envelope with PCM
wallboards: Design tool development, Renew. Sustain. Energy Rev. 31 (2014)
554–562, http://dx.doi.org/10.1016/j.rser.2013.12.031 URL http://www.
sciencedirect.com/science/article/pii/S1364032113008459 .
[13] Y. Kusama, Y. Ishidoya, Thermal effects of a novel phase change material (PCM)
plaster under different insulation and heating scenarios, Energy Build. 141 (2017)
226–237, http://dx.doi.org/10.1016/j.enbuild.2017.02.033 URL http://www.
sciencedirect.com/science/article/pii/S0378778816317923 .
[14] E. Rodriguez-Ubinas, B.A. Arranz, S.V. Snchez, F.N. Gonzlez, Influence of the use of
PCM drywall and the fenestration in building retrofitting, Energy Build. 65 (2013)
464–476, http://dx.doi.org/10.1016/j.enbuild.2013.06.023 URL http://www.
sciencedirect.com/science/article/pii/S0378778813003617 .
[15] M. Saffari, A. de Gracia, S. Ushak, L.F. Cabeza, Passive cooling of buildings with
phase change materials using whole-building energy simulation tools: A review,
Renew. Sustain. Energy Rev. 80 (2017) 1239–1255, http://dx.doi.org/10.1016/j.
rser.2017.05.139 URL http://www.sciencedirect.com/science/article/pii/
S136403211730758X .
[16] N. Soares, J. Costa, A. Gaspar, P. Santos, Review of passive PCM latent heat thermal
energy storage systems towards buildings energy efficiency, Energy Build. 59
(2013) 82–103, http://dx.doi.org/10.1016/j.enbuild.2012.12.042 URL http://
www.sciencedirect.com/science/article/pii/S0378778813000157 .
[17] P.C. Tabares-Velasco, C. Christensen, M. Bianchi, Verification and validation of
energyplus phase change material model for opaque wall assemblies, Build.
Environ. 54 (2012) 186–196, http://dx.doi.org/10.1016/j.buildenv.2012.02.019
URL http://www.sciencedirect.com/science/article/pii/S0360132312000583 .
[18] G. Panayiotou, S. Kalogirou, S. Tassou, Evaluation of the application of phase
change materials (PCM) on the envelope of a typical dwelling in the mediterranean
region, Renewable Energy 97 (2016) 24–32, http://dx.doi.org/10.1016/j.renene.
2016.05.043 URL http://www.sciencedirect.com/science/article/pii/
S0960148116304542 .
[19] S. Ramakrishnan, X. Wang, J. Sanjayan, J. Wilson, Thermal performance of build-
ings integrated with phase change materials to reduce heat stress risks during ex-
treme heatwave events, Appl. Energy 194 (2017) 410–421, http://dx.doi.org/10.
1016/j.apenergy.2016.04.084 URL http://www.sciencedirect.com/science/
article/pii/S0306261916305451 .
[20] P. Marin, M. Saffari, A. de Gracia, X. Zhu, M.M. Farid, L.F. Cabeza, S. Ushak, Energy
savings due to the use of PCM for relocatable lightweight buildings passive heating
and cooling in different weather conditions, Energy Build. 129 (2016) 274–283,
http://dx.doi.org/10.1016/j.enbuild.2016.08.007 URL http://www.sciencedirect.
com/science/article/pii/S0378778816306855 .
[21] A.A. Chernousov, B.Y. Chan, Numerical simulation of thermal mass enhanced
T. Venegas et al.
envelopes for office buildings in subtropical climate zones, Energy Build. 118
(2016) 214–225, http://dx.doi.org/10.1016/j.enbuild.2016.02.054 URL http://
www.sciencedirect.com/science/article/pii/S0378778816301128 .
[22] L. Cabeza, A. Castell, C. Barreneche, A. de Gracia, A. Fernndez, Materials used as
PCM in thermal energy storage in buildings: A review, Renew. Sustain. Energy Rev.
15 (3) (2011) 1675–1695, http://dx.doi.org/10.1016/j.rser.2010.11.018 URL
http://www.sciencedirect.com/science/article/pii/S1364032110003874 .
[23] C. Vlez, M. Khayet, J.O. de Zrate, Temperature-dependent thermal properties of
solid/liquid phase change even-numbered n-alkanes: n-hexadecane, n-octadecane
and n-eicosane, Appl. Energy 143 (2015) 383–394, http://dx.doi.org/10.1016/j.
apenergy.2015.01.054 URL http://www.sciencedirect.com/science/article/pii/
S0306261915000707 .
[24] N. Fumo, P. Mago, R. Luck, Methodology to estimate building energy consumption
using energyplus benchmark models, Energy Build. 42 (12) (2010) 2331–2337,
http://dx.doi.org/10.1016/j.enbuild.2010.07.027 URL http://www.sciencedirect.
com/science/article/pii/S0378778810002574 .
89
Applied Thermal Engineering 141 (2018) 79–89
[25] A.P. Gomes, H.A. de Souza, A. Tribess, Impact of thermal bridging on the perfor-
mance of buildings using light steel framing in brazil, Appl. Therm. Eng. 52 (1)
(2013) 84–89, http://dx.doi.org/10.1016/j.applthermaleng.2012.11.015 URL
http://www.sciencedirect.com/science/article/pii/S1359431112007284 .
[26] South america wmo region 3 – chile, https://energyplus.net/weather-region/south_
america_wmo_region_3, accessed: 2018-04-05.
[27] N/A, American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Inc., 2008.
[28] M. Yazdanian, J. Klems, Measurement of the exterior convective film coefficient for
windows in low-rise buildings, ASHRAE Trans. 100 (1) (1994) 1087–1096.
[29] U.S. Department of Energy, EnergyPlus Engineering Reference, 8th ed., 2018.
[30] M. Mirsadeghi, D. Cstola, B. Blocken, J. Hensen, Review of external convective heat
transfer coefficient models in building energy simulation programs: Implementation
and uncertainty, Appl. Therm. Eng. 56 (12) (2013) 134–151, http://dx.doi.org/10.
1016/j.applthermaleng.2013.03.003 URL http://www.sciencedirect.com/science/
article/pii/S1359431113001543 .